KhalilMrini / LAL-Parser

Neural Adobe-UCSD Parser, the current State of the Art in Constituency and Dependency Parsing.
138 stars 24 forks source link

A question to the model output. #24

Open Creling opened 2 years ago

Creling commented 2 years ago

The model will output a (n+1) * (n+1) matrix given an input sentence with n words. Could you kindly explain what does the matrix represent and from where does the +1 come?

image

Monstarrr commented 2 years ago

In my opinion, +1 come from the addition of node "ROOT", for it is a joint representation of both dependency tree and constituency tree.